From owner-freebsd-questions  Tue Jul  1 10:07:19 1997
Return-Path: <owner-questions>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id KAA08733
          for questions-outgoing; Tue, 1 Jul 1997 10:07:19 -0700 (PDT)
Received: (from jmb@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id KAA08719;
          Tue, 1 Jul 1997 10:07:14 -0700 (PDT)
From: "Jonathan M. Bresler" <jmb>
Message-Id: <199707011707.KAA08719@hub.freebsd.org>
Subject: Re: Benchmarks/comparisions between PC->Sparc
To: scrappy@hub.org (Marc G. Fournier)
Date: Tue, 1 Jul 1997 10:07:14 -0700 (PDT)
Cc: questions@freebsd.org
In-Reply-To: <Pine.NEB.3.95.970701121533.20086D-100000@hub.org> from "Marc G. Fournier" at Jul 1, 97 12:16:27 pm
X-Mailer: ELM [version 2.4 PL24]
Content-Type: text
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Marc G. Fournier wrote:
> 
> 
> Hi...
> 
> 	Does anyone know of URLs for comparisons between the two?  Stuff
> like SPECint and anything else that would be pertinent to a discussion of
> the merits of either/or?
	
	there are a number of problems with SPEC or any other fixed
	task benchmark.   the alternative is a dynamic benchmark that 
	starts with a small dataset and grows till the machine is
	totally tanked.

	take a look at http://www.scl.ameslab.gov/scl/HINT/HINT.html

	the benchmark is a test of cpu/cache/memoery speed.  the ouput
	graph clearly shows the effects of using each to store the
	data set for the benchmark.

	i have collected results for a number of machines which i can 
	send you.

	the ames website (listed above) can graph the results for a 
	much larger variety of machines, thou i have better coverage of
	pc's and sun's.
jmb